CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "data=np.load('data.npy')\n",
    "target=np.load('target.npy')\n",
    "targets=np.load('targets.npy')\n",
    "\n",
    "#loading the save numpy arrays in the previous code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       ...,\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense,Activation,Flatten,Dropout\n",
    "from tensorflow.keras.layers import Conv2D,MaxPooling2D\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1238, 150, 150, 1), (138, 150, 150, 1), (1238, 2), (138, 2))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_data,test_data,train_target,test_target=train_test_split(data,target,test_size=0.1)\n",
    "\n",
    "train_data.shape,test_data.shape,train_target.shape,test_target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_data,test_data,train_target,test_target=train_test_split(data,target,test_size=0.1)\n",
    "\n",
    "train_data.shape,test_data.shape,train_target.shape,test_target.shape\n",
    "\n",
    "model = Sequential([\n",
    "    \n",
    "    #The first CNN layer followed by Relu and MaxPooling layers\n",
    "    Conv2D(100, (3,3), activation='relu', input_shape=(150, 150, 1)),\n",
    "    MaxPooling2D(2,2),\n",
    "    \n",
    "    #The second convolution layer followed by Relu and MaxPooling layers\n",
    "    Conv2D(200, (3,3), activation='relu'),\n",
    "    MaxPooling2D(2,2),\n",
    "    \n",
    "    Conv2D(300, (3,3), activation='relu'),\n",
    "    MaxPooling2D(2,2),\n",
    "    \n",
    "    #Flatten layer to stack the output convolutions from second convolution layer\n",
    "    Flatten(),\n",
    "    Dropout(0.5),\n",
    "    Dense(50, activation='relu'),\n",
    "    Dense(2, activation='softmax')\n",
    "])\n",
    "#The Final layer with two outputs for two categories\n",
    "\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-6-0b93b89907cb>:9: Model.fit_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use Model.fit, which supports generators.\n",
      "WARNING:tensorflow:sample_weight modes were coerced from\n",
      "  ...\n",
      "    to  \n",
      "  ['...']\n",
      "Train for 20 steps, validate on 138 samples\n",
      "Epoch 1/10\n",
      "20/20 [==============================] - 150s 7s/step - loss: 0.7378 - acc: 0.5299 - val_loss: 0.6490 - val_acc: 0.6812\n",
      "Epoch 2/10\n",
      "20/20 [==============================] - 175s 9s/step - loss: 0.6279 - acc: 0.6583 - val_loss: 0.5295 - val_acc: 0.7899\n",
      "Epoch 3/10\n",
      "20/20 [==============================] - 155s 8s/step - loss: 0.5035 - acc: 0.7617 - val_loss: 0.3342 - val_acc: 0.8478\n",
      "Epoch 4/10\n",
      "20/20 [==============================] - 160s 8s/step - loss: 0.3835 - acc: 0.8368 - val_loss: 0.2949 - val_acc: 0.8841\n",
      "Epoch 5/10\n",
      "20/20 [==============================] - 163s 8s/step - loss: 0.3094 - acc: 0.8780 - val_loss: 0.1785 - val_acc: 0.9565\n",
      "Epoch 6/10\n",
      "20/20 [==============================] - 165s 8s/step - loss: 0.2240 - acc: 0.9144 - val_loss: 0.1568 - val_acc: 0.9420\n",
      "Epoch 7/10\n",
      "20/20 [==============================] - 172s 9s/step - loss: 0.2000 - acc: 0.9370 - val_loss: 0.0953 - val_acc: 0.9638\n",
      "Epoch 8/10\n",
      "20/20 [==============================] - 173s 9s/step - loss: 0.1684 - acc: 0.9426 - val_loss: 0.0891 - val_acc: 0.9710\n",
      "Epoch 9/10\n",
      "20/20 [==============================] - 173s 9s/step - loss: 0.1583 - acc: 0.9443 - val_loss: 0.1129 - val_acc: 0.9710\n",
      "Epoch 10/10\n",
      "20/20 [==============================] - 154s 8s/step - loss: 0.1566 - acc: 0.9402 - val_loss: 0.0859 - val_acc: 0.9565\n"
     ]
    }
   ],
   "source": [
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "# create data generator\n",
    "datagen = ImageDataGenerator(width_shift_range=0.1, height_shift_range=0.1, horizontal_flip=True)\n",
    "# prepare iterator\n",
    "it_train = datagen.flow(train_data, train_target, batch_size=64)\n",
    "# fit model\n",
    "steps = int(train_data.shape[0] / 64)\n",
    "history = model.fit_generator(it_train,steps_per_epoch=20,epochs=10,validation_data=(test_data,test_target),verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.plot(history.history['loss'],'r',label='training loss')\n",
    "plt.plot(history.history['val_loss'],label='validation loss')\n",
    "plt.xlabel('# epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['acc'],'r',label='training accuracy')\n",
    "plt.plot(history.history['val_acc'],label='validation accuracy')\n",
    "plt.xlabel('# epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "138/138 [==============================] - 5s 33ms/sample - loss: 0.0859 - acc: 0.9565\n",
      "[0.08594163204880728, 0.95652175]\n"
     ]
    }
   ],
   "source": [
    "print(model.evaluate(test_data,test_target))\n",
    "model.save('model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_classes = model.predict(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_classes = np.argmax(np.round(predicted_classes),axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((138,), (138, 2))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predicted_classes.shape, test_target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.],\n",
       "       [1., 0.],\n",
       "       [1., 0.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_target"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}